Forested buffer strips are used to mitigate fragmentation and habitat loss and are a common feature in management of riparian landscapes. Low-elevation, old-growth coastal forests are a rare riparian habitat that can benefit from similar conservation measures. We evaluated the effectiveness of postlogging, forested buffer strips for forest-dwelling birds in the coastal temperate rainforest of southeast Alaska, USA. Our objective was to compare bird composition and density among forested buffer strips of differing widths and controls at the stand and landscape scales. We applied a 2-stage sampling design stratified by forested buffer width and randomly selected 24 managed and 18 control sites to sample over 2 breeding seasons. We estimated abundance of birds using the paired-observer, variable-circular plot method. We modeled combined effects of buffer width and vegetation and landscape characteristics on bird density at 2 spatial scales. Species richness and diversity were greatest in the narrowest buffers, but species composition in the largest buffers (≥400 m) was most similar to that in control blocks. Abundance of 3 of 10 common species differed across forested buffer treatments and controls. Densities of red-breasted sapsucker (Sphyrapicus ruber) and Pacific-slope flycatcher (Empidonax difficilis) were positively related to buffer width, whereas density of ruby-crowned kinglet (Regulus calendula) was negatively associated with buffer width. Parameter estimates for buffer width effects at both spatial scales were similar within species. We found few habitat and landscape variables that clearly affected density of our focal species, and among species no predictor variables affected density in a similar fashion. We recommend retaining forested buffers ≥400 m to support composition and abundance of forest-dwelling birds, particularly those species that rely on interior forest conditions.